Share product ideas and upvotes with our product team
Hello Intercom Community!I’ve been building complex automated shops using Fin and Workflows, and I’ve noticed a significant limitation in the current "Let Fin handle" node. Once a user is inside the Fin flow, it often feels like a "dead end" for automation. Currently, Fin can only resolve the chat or escalate to a teammate.The Problem: There is currently no built-in way for Fin to guide users back to previous paths or nodes within the same conversation flow once the AI is active. For example, if a user finishes browsing product details with Fin but wants to go back to the main menu buttons to start a different process, they are stuck unless a human intervenes.The Proposed Feature: "Automation Loop-Back" I would love to see an option within Fin's guidance or node settings that allows for a "Start Over" or "Go Back" trigger.Key Functionality: A specific command in Fin's guidance (e.g., "If the user wants to start over, hand back to [Node Name]") that re-triggers a previous part of the workflow.Why it matters: This would make Fin truly modular. It would allow us to build "circular" experiences where users can bounce between AI-powered information gathering and structured, robot-led data collection (like order forms) without the conversation feeling broken.Current Workaround (The Tip): Until this is a native feature, I’ve found that the best "fix" is to use Handover Paths connected to Reusable Workflows. By forcing Fin to "Hand over" after a specific intent, you can land the user back into a fresh workflow node that presents the menu again.Has anyone else found a better way to let users "navigate" backward while talking to Fin?I’d love to hear your thoughts!
Our database is not linked to Intercom and we cannot upload a list of users because we deal with PHI so need to limit contact information to those we correspond with, so ALL contacts are leads. We have many, many duplicate leads and need to be able to merge them so all of their conversations (phone, emails, tickets, etc) are together.
The problem is two fold, this ticket focuses on the second point below:Intercom only allows the merge of duplicate leads into users (which will get it’s own ticket) Currently a lead is only changed to a user if the system detects an account has been created, or if we import a list of users. We do not currently have our database linked to Intercom and deal with PHI so it is important we keep data sharing to as needed, so importing a full list isn’t ideal.We should be able to manually change leads to users so that we can then merge duplicate leads into the matching user.
Im looking to understand the performance and utilization of help centre article so i can understand how they’re performing. Also to be able to segment views by the customer data I’m flowing into Intercom. Therefore, similar to how we’re able to create customizable reports to review Conversation and Ticket data, I’d also like to be able to review Article data in this way. The Legacy report is a starting point but doesn’t have enough customization or filter options to make it valuable to understand how well the information is being utilized by the intended audience/s.
Would be great to be able to color code inbox by statuses, so you can quickly view an inbox and see the status of all items.
As can propose Zendesk, it would be more efficient and easier to organize the macros, if we were able to create a single macro, for different languages.
Please can some improvements be made to the Help Centre’s search functionality? Currently it isn’t really up to industry standard and is providing a poor user experience• It doesn’t pick on synonyms/different terminology - so without keyword-stuffing (which is bad practice and poor UX), content gets missed. Already raised by someone else here, here and here a year ago and still an issue - especially when it comes to typos, Fin can detect these so the HC should be able to as well. Some typos appear to be accounted for but not all - we should be able to customise this.• Similarly, if we have a standardised/approved way of spelling/styling a term, this is the only one we use within our content for consistency - however users might spell it differently. Their spelling is valid (and not a typo necessarily) but doesn’t match our brand-approved spelling, and so articles don’t surface for them. Eg. “multisite”, “multi-site”, “multi site” are all grammatically appropriate, but we only use the hyphenated one according to our brand guidelines - if a user searches “multisite”, the articles don’t show in the search results • There is no way to filter (eg. by topic, region) - much of our content is regionalised, but there’s no way to distinguish this without using the article descriptions which aren’t always visible in search results (as it previews the section where the keyword appears). Users have to open and check the description to know if it’s relevant, which is a huge time-waster. Users should be able to find which responses are relevant to them, just like Fin regionalises if you use audiences.• Certain articles need to be visible in collections for visibility (and therefore can’t be “unlisted”) but are not applicable in searches, and are taking up space in those searches, stopping more relevant content coming to the top. For example our historic release notes currently live in the Help Centre for visibility, but entries from 2018 shouldn’t be surfaced in searchesPossible solutions• Transparent keywords (as suggested here) to indicate topics covered in an article without needing to include in copy • Tags to achieve the same as above but also allow for filtering (eg. search only articles relevant to AU users, or only articles on a particular topic)• Toggle to exclude from searches (similar to the toggle to exclude from suggested articles) - would ensure they don’t clog up search results, without having to hide from the Help Centre
Many companies are now providing llms.txt for their documentation to make it more efficient for agents to consume content. It’s an index of publicly available markdown files. See, for example, https://cloudinary.com/documentation/llms.txt. If Fin supported this as a content source (or even just markdown files as a source) it would really help us, as the markdown files contain all code examples, and expand content that’s in collapsed sections - content that’s not accessible directly from the HTML.
Ability to define and save a company style and tone guide. As employees compose messages in chat or email, provide real-time, AI-powered coaching to help align with that guide. Before reps press send, run a final check against our tone and style guides and prevent delivery if the message doesn’t meet standards, with clear guidance on what needs to be updated.
It would be good to be able to like/dislike and correct fin responses in real time. There is the “improve answer” option, but I would like to directly tell fin that the answer it gave was incorrect or misleading, and course correct for future responses. If it gave a good response, I would like to indicate that somehow. I think this would help a lot with training it outside of help centre materials
FIN AI should be able to monitor how many tickets come in in a given timeframe. If this is suddenly spiking, and especially if the same topic is mentioned in multiple tickets, then FIN AI should escalate directly to human support with high priority. Plus spark and alert.
We need a feature to detect Fin attributes / ai categories even if in a Workflow or Channel (even if Fin isn’t enabled for that workflow/channel)Previously in the AI Categorization Beta there was a feature to add AI Category detection to a workflow so we had more control over when Fin set the attributes. A few use cases that would benefit from this: Detecting category earlier in the process so we can route conversations using categories and not wait until handover or closure, it seems that Fin doesn’t take this action early enough in some cases to route this effectively. Detecting Spam on Email without Fin over Email being required, this was a very useful case for many clients who may not want Fin on email. Minimizing resolution costs while still tagging attributes - this could be a lower cost for adding attributes or maybe a monthly fee for attribute tagging on non active Fin workflows? See this thread for more context as well
In working with clients, we’ve been running into scenarios where we are trying to improve the response rate and customize Fin’s wording for the end of an assumed resolution to engage the customer and improve the CX score (the client has noticed that CX score is lower if the customer doesn’t confirm the resolution). It would be helpful to be able to prompt this in some way similar to how you can customize other areas of Fin for example in this Follow up with Inactive Customers screen.It would also be helpful to know if CX score is impacted by this and if it is, be able to A/B test this ending with a few options proposed by individual teams.
Attualmente, Intercom suddivide i messaggi automatici (outbound) inviati allo stesso cliente in più conversazioni separate, senza offrire la possibilità di riunirle in un’unica vista.Questo comportamento genera un problema concreto e quotidiano, soprattutto per chi – come noi – utilizza WhatsApp come canale principale di comunicazione con i clienti/pazienti.👉 Quando un cliente risponde a un messaggio automatico, la sua risposta finisce in una conversazione separata.👉 Di conseguenza, la cronologia dei messaggi risulta frammentata, rendendo difficile: ricostruire il contesto trovare rapidamente un messaggio ricevuto avere una visione chiara e continua della relazione con il cliente Nel nostro caso (ambito sanitario), vedere la conversazione con un paziente spezzettata in più thread è particolarmente deleterio: aumenta il rischio di errori, rallenta l’operatività e peggiora l’esperienza sia per il team sia per il paziente.La richiestaChiediamo la possibilità di unire tutti i messaggi in entrata e in uscita —sia manuali sia automatici (outbound) —sotto un’unica conversazione per cliente, indipendentemente dal tipo di messaggio o dal trigger che lo ha generato.Il valore Una single source of truth per ogni cliente Migliore continuità conversazionale su WhatsApp Meno tempo perso a cercare messaggi tra conversazioni spezzate Migliore qualità del servizio e maggiore sicurezza operativa Questa funzionalità migliorerebbe in modo significativo l’esperienza di tutti i team che utilizzano Intercom per comunicazioni strutturate e ad alto volume, specialmente su canali di messaggistica come WhatsApp.
At the moment, when starting a new conversation it fills the bottom half of the screen. When on a laptop or just working from one screen, it’s incredibly annoying. Can we shift the alignment so the conversation is higher up on the page? The scroll feature does nothing except to show shortcuts.
Problem: Currently, the workflow creates a task in Jira and sends a notification to Slack, but the notification only contains a link to the Intercom object/conversation. This forces moderators to manually search for the created issue in Jira, which slows down work.What you need:Add a variable with a URL to the created Jira issue (or at least the issue key + a clickable link) to the Slack notification. If you can't insert it into the first message, send a second reply to the thread of the same Slack notification.Why/Value: Switch to Jira faster, less manual work, fewer copy/paste/search errors.
Quick heads-up for anyone using the “customer clicks a website element” trigger 👋Right now, if you make changes to a workflow—like renaming a chatbot or adjusting the workflow—it’s expected that you may need to re-select and re-save the trigger again.This happens because the trigger is tied to a specific website element via a CSS selector, and certain changes can break that link behind the scenes. If the element isn’t re-selected, the workflow may not fire as expected.While this is currently a known limitation (not a bug), it would be great to see an improvement here 🙏Feature request: preserve the element selection automatically when workflows are renamed or updated, so we don’t have to manually re-save each workflow after every change.If this would save you time (and a bit of head-scratching 😅), please upvote! Would also love to hear how often this impacts your setup.
Hey team, I think it would great if we could favorite or star useful articles that we come back to time and time again. Right now, I’m having to save them all in a bookmark folder through my browser, but I want to be able to see my most frequented articles within the help center.Would be great to have it show frequently visited/recent articles too!Our customers would love this as well!Thanks!Emily
It would be great if we could add temporary collaborators to a document or project in Intercom as not all our colleagues have access to Intercom.Individual engineering or product team members work on a new feature / function which we then want to document. Previously we would draft everything in our internal repository, however when copying over into IC, images, etc do not come over and have to be added back in manually. So currently we use the option to ‘publish internally’ (no collection / FIN turned off) which allows us to draft directly in Intercom and then provide our SME’s with a link to the draft copy for review. However, this then creates a feedback loop issue. For the reviewer they need to copy / paste or draft notes externally, etc…. It’s cumbersome and creates a bit of friction in the feedback process. Suggestion would be to offer the ability to give users a temporary collaborator link on a specific article url (could be used for other features as well) which then allows them to add comments / make edits (permission control option on link?) directly in the page.
Already have an account? Login
No account yet? Create an account
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.